Peft Python Packages

peft

🤗 PEFT: State-of-the-art Parameter-Efficient Fine-Tuning.

ms-swift

Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 600+ LLMs (Qwen3.8, DeepSeek-V4, GLM-5.1, InternLM3, Llama4, ...) and 300+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, GLM5.3, Gemma4, Llava, Phi4, ...) (AAAI 2025).

gptqmodel

LLM model quantization (compression) toolkit with HW acceleration support for Nvidia, AMD, Intel GPU and Intel/AMD/Apple CPU via HF, vLLM, and SGLang.

llamafactory

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

soup-cli

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

mcore-bridge

MCore-Bridge: Providing Megatron-Core model definitions for state-of-the-art large models and making Megatron training as simple as Transformers — with support for 300+ large language models (Qwen3-Next, GLM-5.3, Deepseek-V4, MiniMax-2.7, ...) and 200+ multimodal large models (Qwen3.8, Qwen3-Omni, Gemma4, ...).

trloom

YAML-driven end-to-end fine-tuning on Hugging Face TRL — one config for model, dataset, W&B, and Modal GPUs.

miniverl

verl for a single consumer GPU. PPO, GRPO and on-policy distillation on NVIDIA GPUs.

mlx-tune

Fine-tune LLMs on your Mac with Apple Silicon. SFT, DPO, GRPO, Vision, TTS, STT, Embedding, and OCR fine-tuning — natively on MLX. Unsloth-compatible API.

abliterix

Automated alignment adjustment for LLMs — direct steering, LoRA, and MoE expert-granular abliteration, optimized via multi-objective Optuna TPE.

thekaveh-nnx

Lightweight PyTorch toolkit for training, fine-tuning, and exporting modern neural nets. FFN, GNN, decoder-only LM, diffusion, JEPA, MoE, PEFT, quantization, pruning, surgery, ONNX and Hugging Face Hub interop, plus experimental GGUF/Ollama bundle export.

canifinetune

Single-GPU LLM fine-tuning preflight: memory estimates, runnable LoRA/QLoRA recipes, and local measurements.

block-streamer

Stream PyTorch model blocks from NVMe or pinned CPU memory with a bounded GPU residency budget. LoRA-finetune models far larger than host RAM on one GPU - Qwen3-VL-32B and gpt-oss-120b train in under 6 GB of VRAM.

polywhisper

🎙️ PolyWhisper v9: efficient multilingual Indic ASR — frozen Whisper-Small + per-language LoRA for Hindi, Tamil, Telugu, Bengali, Marathi. Bengali −34.5%, Marathi −43.2% vs baseline. ONNX CPU-ready. Apache-2.0.

llmtuner

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

maatml

Turn a multi-gigabyte base model into a small subject-matter expert you can run anywhere. One model.yml drives prepare, train, evaluate, export, and serve, with validators checking correctness outside the model at each stage.

forgelm

Config-driven LLM fine-tuning with safety evaluation, EU AI Act compliance, 6 alignment methods, and one-command bundled quickstart templates.

clef-finetune

Fine-tune Cloudflare's open-source Clef / Clef-flash decision models (LoRA + joint schema head) for vertical decisions like insurance claims and compliance

shadowlm

A fine-tuning SDK — any open model, any harness, any method. 12 training methods behind one argument; pure-stdlib core.

ownvoice-cli

CLI that trains a LoRA voice-cloning adapter for pocket-tts TTS, saved to disk, not an API. PyPI + npm.

xturing

Build, personalize and control your own LLMs. From data pre-processing to fine-tuning, xTuring provides an easy way to personalize open-source LLMs. Join our discord community: https://discord.gg/TgHXuSJEk6

lazyllm-llamafactory

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

llm-gpu-lab

One GPU. Full LLM workflow. Real benchmarks. No cloud required.

unsloth-mlx

Fine-tune LLMs on your Mac with Apple Silicon. SFT, DPO, GRPO, Vision, TTS, STT, Embedding, and OCR fine-tuning — natively on MLX. Unsloth-compatible API.